Effect of combined catheter ablation of atrial fibrillation and left atrial appendage closure on left atrial structure compared with a single procedure

To the Editor: Atrial fibrillation (AF) is a well-recognized cause of left atrial (LA) structural remodeling.^[1] LA sphericity (LASP), a new shape-based remodeling parameter, is an independent predictor for AF ablation outcome.^[2,3] Improvement of LA structure was defined as LA reverse remodeling (RR). This study compared the impact of combined left atrial appendage (LAA) closure (LAAC) and catheter ablation (CA) of AF with a single procedure on LA volume (LAV) and LASP.

We retrospectively enrolled consecutive patients with non-valvular AF between January 2017 and December 2020. The study protocol was approved by the Ethics Committee of Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine (No. NCT03788941). All patients provided informed consent. The inclusion criteria were the following: (1) age ≥18 years; (2) CHA₂DS₂-VASc (congestive heart failure, hypertension, age ≥75 years [doubled], diabetes, stroke [doubled], vascular disease, age 65–74 years, and sex category) Score ≥2; (3) symptomatic AF refractory to antiarrhythmic drugs with CA indication; and (4) LAAC with at least one of the following indications: HAS-BLED (hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile international normalized ratio, elderly, drugs/alcohol concomitantly) score ≥3; thromboembolism (TE) even with oral anticoagulants (OAC); and contraindications or unwillingness to receive long-term OAC. We excluded patients with severe valvular heart disease, hyperthyroidism, or incomplete data. Patients were divided into CA-only, LAAC-only, and combined groups. The combined group was 1:1 matched with the other two groups by propensity score matching (PSM), respectively (combined vs. CA-only and combined vs. LAAC-only).

LAA thrombus was ruled out by preoperative transesophageal echocardiography (TEE). CARTO (Biosense Webster, Diamond Bar, CA, USA) or EnSite (St. Jude Medical, St. Paul, MN, USA) electroanatomical mapping systems were used for LA reconstruction and ablation. For patients with paroxysmal AF, pulmonary vein (PV) isolation was performed. Additional linear ablations were subsequently performed for those with persistent AF. Sinus rhythm (SR) was restored by ablation or electric cardioversion. In the combined group, Watchman device (Boston Scientific Corporation, Natick, MA, USA) implantation was performed after CA. Immediate intraprocedural TEE and/or angiography were performed to verify appropriate device implantation.

During follow-up visits performed at 3 months and 6 months postoperatively and every 6 months thereafter, electrocardiography (ECG) and Holter tests were performed to detect AF recurrence. TEE and coronary computed tomography angiography (CCTA) were performed for detecting peri-device leaks (PDL) and device-related thrombus. The OAC strategy for patients who received LAAC was adjusted according to the satisfactory seal results.

LAV and LAA volume (LAAV) were measured by Mimics Medical 17.0 (Materialise NV, Leuven, Belgium), a software tool for visualizing and segmenting medical images and rendering 3D objects [Figures 1A and 1B]. Separation of the LA from the left ventricle, PVs, and LAA was bounded by the mitral valve annulus, PV ostia, and LAA orifice, respectively. LAV and LAAV were automatically calculated using the Mimics software (Materialise NV). Original CCTA images were reconstructed for chamber measurements using Extended Brilliance Workspace version 4.5 (Philips Healthcare Cleveland, OH, USA). From three diameters of the LA, including the largest transverse and anterior-posterior diameters measured on axial planes and the largest craniocaudal diameter on the sagittal plane, the maximum was selected for calculating LASP [Figures 1C and 1D].

Figure 1.

Three-dimensional images of the LA (yellow) and LAA (blue-green) (A) and separated LA (B), as well as the measurement of largest LA transverse, anterior-posterior diameters on the axial plane (C) and craniocaudal diameter on the sagittal plane (D). The volume of the given LA is 77.13 cm³ and 6.95 cm is chosen as the maximum diameter. The diameter of the sphere with the same volume of 77.13 cm³ is 5.28 cm. From these values, the LASP is calculated as 75.97% (5.28/6.95 = 0.7597). LA: Left atrial; LAA: Left atrial appendage; LASP: LA sphericity.

Statistical analysis was performed using SPSS 26.0 (IBM Software, Armonk, NY, USA). Two-sample t-test, paired t-test, or Mann–Whitney U test was used to compare continuous variables, and the Pearson chi-squared test or Fisher's exact test was used to compare categorical data as appropriate. Survival data were analyzed using the Kaplan–Meier method with a log-rank test. Multiple linear/logistic regression analyses were performed to identity the clinical factors associated with outcomes. A P <0.05 was considered statistically significant.

There were 252, 157, and 63 patients meeting the criteria in the combined, CA-only, and LAAC-only groups, respectively. CA-only and LAAC-only groups showed significant differences in several clinical characteristics with the combined group, respectively. After PSM, 141 and 52 pairs were successfully matched in the combined group vs. CA-only group and combined vs. LAAC-only group, respectively. All characteristics were comparable at baseline, except for the HAS-BLED score in the combined group vs. CA-only group (3.1 ± 1.2 vs. 1.6 ± 1.0, P <0.001). During the procedure, SR was restored in all patients who underwent CA. Watchman devices (Boston Scientific Corporation) were successfully implanted with a 100% satisfactory seal in patients who received LAAC. No major bleeding occurred during hospitalization, and all patients were switched to OAC before discharge.

One year after the intervention, SR was maintained in 78.2% (197/252) and 75.8% (119/157) of patients in the combined and CA-only groups, respectively. There were four patients with TE, one with major bleeding (MB) in both the combined and CA-only groups, one patient with TE, and one with MB in the LAAC-only group. No device-related thrombus or major PDL was detected on the follow-up imaging examination.

In the combined and CA-only groups, LAV significantly decreased from 128.38 ± 41.37 mL to 111.79 ± 35.17 mL (P <0.001) and 114.80 ± 43.65 mL to 99.39 ± 38.88 mL (P <0.001) after therapy, respectively. A significant LAV increase was noticed in the LAAC-only group (175.98 ± 49.61 mL vs. 185.96 ± 55.96 mL; P <0.001). LASP significantly increased after treatment in both the combined ([79.60 ± 5.25]% vs. [80.90 ± 5.74]%, P <0.001) and CA-only groups ([80.43 ± 5.92]% vs. [82.89 ± 5.56]%, P <0.001), but not in the LAAC-only group ([81.60 ± 5.37]% vs. [82.33 ± 5.30]%, P = 0.221). After PSM, LAV changes (△-volume) were similar in the combined and CA-only comparison (-13.52 ± 22.72 mL vs. -16.63 ± 24.59 mL; P = 0.250), but differed significantly between the combined and LAAC-only comparison (-21.68 ± 22.22 mL vs. 9.11 ± 19.69 mL; P <0.001). Post-procedural LASP changes (△-sphericity) differed significantly between the combined and CA-only groups ([1.42 ± 4.40]% vs. [2.54 ± 4.72]%, P = 0.028), but were similar between the combined and LAAC-only groups ([1.46 ± 4.10]% vs. [0.68 ± 5.00]%, P = 0.346).

Combined application of LAAC and CA was positively correlated with spherical RR (odds ratio [OR], 1.66; 95% confidence interval [CI], 1.009–2.716; P = 0.045) when taking CA as a reference. Compared with patients (92, 36.5%) exhibiting both volumetric and spherical RR, those patients (44, 17.5%) with neither volumetric nor spherical RR had a higher major adverse cardiovascular event (MACE) rate (P <0.05). Multiple logistic analysis showed that in the combined group, RR were independently associated with the TE events (OR, 0.067; 95% CI, 0.007–0.658; P <0.020), but the AF recurrence (OR, 1.570; 95% CI, 0.162–15.188; P = 0.697), age (OR, 1.082; 95% CI, 0.921–1.271; P = 0.341), or other characteristics were not associated with TE events.

Preoperative LASP (B = 0.335; P <0.001), △-volume (B = 0.055; P <0.001), and combined LAAC (B = 2.373; P <0.001) interactively affected the extent of spherical RR after ablation and explained 49.0% of its variation (adjusted R² = 0.490). The multiple linear regression equation could be written as:

$$Y=-\mathrm{30.259}+\mathrm{0.335}\times \mathrm{p}\mathrm{r}\mathrm{e}\mathrm{o}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{L}\mathrm{A}\mathrm{S}\mathrm{P}+\mathrm{0.055}\times$$

△– $\mathrm{v}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e}+\mathrm{2.373}\times \mathrm{c}\mathrm{o}\mathrm{m}\mathrm{b}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{L}\mathrm{A}\mathrm{A}\mathrm{C}$

where Y = extent of LA spherical RR (%);preoperative LASP = LASP (%); △–volume = post-ablation LAV changes (mL);and combining LAAC = 1, otherwise = 0 in the equation.

We found that the addition of LAAC to the CA procedure did not affect the improvement of LAV after CA but could alleviate spheroidization from the ablation scar. Compared to LAAC alone, combined LAAC and CA resulted in a significant reduction in LAV.

LA structural changes after CA-only could be explained by scar-induced retraction. Bisbal et al^[1] believed that post-ablation RR is caused by scarring and myocardial structural recovery. Real RR was presented as reductions in both volume and sphericity, and the scar contraction caused by ablation played a more important role in volumetric RR only. The reduced volume after ablation might mainly locate at the PVs ridge junction to the LA, leading to a more symmetric structure of the LA. Gottlieb et al^[4] discovered that the diameter of the ablated PV decreased similarly in patients with AF and healthy sheep, further supporting this hypothesis.

We noticed that the probability of spherical RR after CA was 0.66 times higher than the combination of LAAC and CA. The Watchman device limited the volume reduction in the LAA ostium, which increased the irregularity of the LA. The following three structural changes in the LA may occur after different treatments: reduction in both volume and sphericity caused by real RR; volume reduction and sphericity increase caused by ablation scar; and increase in both volume and sphericity as a result of remodeling progression. Both combined therapy and LAAC alone resulted in increased sphericity through different mechanisms, leading to no significant difference in △–sphericity. And, real RR was associated with better cardiovascular outcomes after combined procedures.

The symmetric structure generated less vortical flow and more areas with slow flow and stasis, facilitating the formation of thrombi, which supports the value of the LASP for non-LAA thrombogenesis risk assessment.^[5] Combining LAAC and CA for AF eliminates LAA thrombosis directly and reduces the probability of cardioembolic stroke from non-LAA thrombosis by easing spheroidization caused by ablation scar.

The study has some limitation. First, this was a single-center retrospective study with a cohort consisting of three different non-randomized groups and the bias may exist. We attempted to minimize the patient selection bias using PSM. Second, the number of patients and the follow-up time in this study were limited, and larger and longer studies are required to confirm these conclusions. Third, LAV and LASP measurements may differ betweem pre- and post-LAAC (sometimes the device protrudes into the LA), and the reduction of LAV and LASP may be overestimated.

Compared with a single procedure, combined CA and LAAC resulted in LAV reduction with less spheroidization. The combination of LAAC, larger preoperative LASP, and more LAV reduction post-procedure were associated with more notable LASP reduction after ablation. Future cohort studies with larger sample size are needed to validate these results.

Funding

This work was supported by grants from the State Key Program of National Natural Science Foundation of China (No. 82130009), the National Science Foundation of China (No. 82070515), and the Clinical Research Plan (No. SHDC2020CR2026B).

Conflicts of interest

None.